1
$\begingroup$

I need to evaluate the sample size of given study (56 samples). An ordinal logistic regression is used to compute a model to find significant effects with three binary independent, and three ordinal numeric dependent variables.

In the study, participants are split into eight groups of seven. Each group is then tested under one of the possible eight conditions of the three binary independent variables.

And I need to assess whether the size is critical/sufficient. Using the rule of thumb (Peduzzi et al. (1996)) I get 30 = 3 * 10, so 56 is a good size?

or do the variables in this case need to counted 10 * (2*3) = 60? Also the independent variables are between-subject variables in this case, does that have an effect on the size?

$\endgroup$
  • $\begingroup$ You could try simulation $\endgroup$ – kjetil b halvorsen Sep 6 '17 at 13:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.